Animals carrying a heterozygous loss-of-function mutation in the Scn1a gene (Scn1a+/- mice), which encodes a subunit of the voltage-gated Na+ channel NaV1.1, show deficits in both homeostatic regulation of sleep and circadian regulation of rest-activity cycles. The NaV1.1 channel is the primary voltage-gated Na+ channel in adult GABAergic interneurons and its reduced activity results in a decrease of GABAergic tone, suggesting that sleep regulatory deficits in Scn1a+/- mice emerge from reduced GABAergic activity. However, Scn1a+/- mice are a model of a severe form of epilepsy known as Dravet syndrome (DS) and, as DS patients, show not only dysregulation of sleep but also generalized seizures. Epileptic activity complicates the interpretation of sleep disorders in DS, as sleep regulatory deficits could be the result of sleep disruption by seizures or of seizure-associated neural damage. Our hypothesis is that sleep disorders in DS are the consequence of reduced GABAergic tone within sleep regulatory centers that is independent of the presence of seizures. To address this hypothesis, we propose to conditionally target the Scn1a+/- mutation to specific neurons and brain regions. Specific Aim 1 will determine whether sleep abnormalities in global Scn1a+/- mice emerge from the effect of the mutation specifically on GABAergic neurons. We will target the Scn1a+/- mutation to these cells using an Scn1alox/- mouse and a Cre driver mouse line that targets GABAergic neurons throughout the brain, and will assess the integrity of the circadian and homeostatic regulation of sleep in these mutants. This approach will unequivocally determine whether sleep regulatory deficits in Scn1a+/- mice are the result of reduced NaV1.1 channel activity within GABAergic cells or whether non- GABAergic cells that express the channel also contribute to this phenotype. Specific Aim 2 will target the Scn1a+/- mutation to cells in the suprachiasmatic nucleus (SCN), the site of the central circadian pacemaker that regulates sleep. Viruses expressing Cre recombinase, targeting either all SCN cells or specifically vasoactive intestinal polypeptide (VIP)-containing cells, will be injected wihin the SCN of Scn1alox/+ mice. We will also target the mutation to the SCN VIPergic cells by crossing Scn1alox/+ mice with a mouse line in which the VIP promoter drives the expression of Cre. Because VIP neurons are essential for the integrity of the SCN oscillatory network, we expect that these VIP-specific Scn1a+/- mutants will show similar effects to mutants in which all SCN cells are targeted. Specific Aim 3 will virally target the Scn1a+/- mutation to the reticular nucleus of the thalamus (RNT), which is essential for the generation of slow-wave sleep and spindles during non- REM sleep, both compromised in Scn1a+/- mice. None of the conditional mutant approaches in Aims 1 and 2 is expected to induce seizures, offering a unique opportunity to assess the effect of reduced NaV1.1 channel activity in sleep regulatory regions, in the absence of seizures. We predict that the conditional targeting of the Scn1a+/- mutation to the SCN and RNT will lead to deficits in circadian and homeostatic regulation of sleep, respectively. These results would provide direct support for the role of the NaV1.1 channel within these brain regions in the regulation of sleep. They would also directly support our hypothesis that both circadian and homeostatic sleep deficits in DS emerge from seizure-independent reduced GABAergic activity in specific brain regions, providing new avenues for the treatment of sleep disorders in DS.